Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/bytedance/flux
/ functions
Functions
784 in github.com/bytedance/flux
⨍
Functions
784
◇
Types & classes
309
Method
gemm_kernel
src/reduce_scatter/gemm_v2_reduce_scatter.hpp:268
Method
gemm_only
src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:637
Method
gemm_only_impl
src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:586
Method
gen_head
include/flux/op_registry.h:48
Method
gen_tail
include/flux/op_registry.h:60
Method
get_all_records
src/ths_op/ths_op.cc:157
Method
get_args_workspace_size
src/reduce_scatter/gemm_v2_reduce_scatter.hpp:472
Method
get_barrier_workspace_size
Get the workspace size needed for barrier
src/all_gather/sm80_all_gather_gemm.hpp:338
Method
get_barrier_workspace_size
src/reduce_scatter/gemm_v2_reduce_scatter.hpp:458
Method
get_barrier_workspace_size
Get the workspace size needed for barrier template <class ThreadblockShape>
src/reduce_scatter/epilogue_evt_nvshmem.hpp:153
Method
get_barrier_workspace_size
Get the workspace size needed for barrier
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:285
Method
get_barrier_workspace_size
src/reduce_scatter/gemm_v3_reduce_scatter.hpp:378
Method
get_barrier_workspace_size
include/flux/gemm_operator_base.h:109
Method
get_barrier_workspace_size
include/flux/cuda/gemm_impls/gemm_operator_base_default_impl.hpp:92
Method
get_block_shape
src/all_gather/sm90_all_gather_gemm_tma_warpspecialized_cooperative.hpp:382
Method
get_block_shape
src/reduce_scatter/sm90_gemm_tma_warpspecialized_cooperative_reduce_scatter.hpp:330
Method
get_callbacks
src/reduce_scatter/epilogue_evt.hpp:386
Method
get_callbacks
src/reduce_scatter/epilogue_evt_nvshmem.hpp:342
Method
get_callbacks
src/reduce_scatter/gemmk_visitor_load.hpp:156
Function
get_consumer_store_callbacks
src/reduce_scatter/sm90_epilogue_evt.hpp:252
Function
get_cu_error_string
include/flux/cuda/cuda_stub.h:50
Method
get_current_work
src/reduce_scatter/tile_scheduler/sm90_tile_scheduler_reduce_scatter.hpp:159
Function
get_dist_env
(deterministic: bool = True)
python/flux/dist_utils.py:71
Method
get_func_attr
src/reduce_scatter/reduce_scatter_kernel.hpp:2175
Method
get_gemm_meta
src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:156
Method
get_grid_blocks
Returns the total number of thread blocks to launch
src/all_gather/sm80_all_gather_gemm.hpp:482
Method
get_grid_blocks
Returns the total number of thread blocks to launch
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:433
Method
get_grid_dims
Returns the grid extents in thread blocks to launch
src/all_gather/sm80_all_gather_gemm.hpp:489
Method
get_grid_dims
Returns the grid extents in thread blocks to launch
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:440
Method
get_grid_shape
Computes the kernel launch grid shape based on runtime parameters
src/all_gather/sm90_all_gather_gemm_tma_warpspecialized_cooperative.hpp:367
Method
get_grid_shape
Computes the kernel launch grid shape based on runtime parameters
src/reduce_scatter/sm90_gemm_tma_warpspecialized_cooperative_reduce_scatter.hpp:315
Function
get_inter_node_rs_group
()
python/flux/gemm_rs_sm80.py:52
Method
get_latest_record
src/ths_op/ths_op.cc:182
Function
get_local_rank_from_env
src/cuda/utils.cc:98
Function
get_local_world_size_from_env
src/cuda/utils.cc:88
Function
get_nvshmem_nodes
(node, types)
pynvshmem/tools/autogen_pybind.py:260
Method
get_optional_scale_tensor
src/comm_none/ths_op/gemm_only.cc:236
Method
get_optional_scale_tensor
src/all_gather/ths_op/all_gather_gemm_kernel.cc:236
Method
get_partials_workspace_size
Get the workspace size needed for intermediate partial sums
src/all_gather/sm80_all_gather_gemm.hpp:349
Method
get_partials_workspace_size
Get the workspace size needed for intermediate partial sums
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:296
Method
get_producer_load_callbacks
src/reduce_scatter/sm90_epilogue_evt.hpp:184
Function
get_rank_from_env
src/cuda/utils.cc:93
Method
get_rt_config
src/all_gather/ths_op/all_gather_gemm_kernel.cc:281
Method
get_rt_config
src/all_gather/ths_op/all_gather_gemm_kernel_crossnode.cc:837
Method
get_runtime_gemm_hparams
include/flux/cuda/gemm_impls/gemm_operator_base_default_impl.hpp:97
Function
get_sm_count
src/cuda/cuda_common.cc:60
Method
get_tile_offset
Obtains the calling threadblock's tiled coordinates for the given tile index
src/reduce_scatter/tile_scheduler/threadblock_swizzle_acrossnode.hpp:78
Method
get_tiled_shape
Returns the GEMM volume in thread block tiles
src/all_gather/sm80_all_gather_gemm.hpp:476
Method
get_tiled_shape
Returns the GEMM volume in thread block tiles
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:427
Method
get_tma_descriptor
Return TmaDescriptor/TensorMap
include/flux/cuda/memory_utils.hpp:388
Function
get_torch_prof_ctx
(do_prof: bool)
python/flux/util.py:113
Method
get_workspace_size
src/all_gather/sm90_all_gather_gemm_tma_warpspecialized_cooperative.hpp:307
Method
get_workspace_size
Returns the workspace size (in bytes) needed for these parameters
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:375
Method
get_workspace_size
src/reduce_scatter/sm90_epilogue_evt.hpp:148
Method
get_workspace_size
src/reduce_scatter/sm90_gemm_tma_warpspecialized_cooperative_reduce_scatter.hpp:255
Method
get_workspace_size
total workspace: args_workspace + gemm workspace
include/flux/gemm_operator_base.h:104
Method
get_workspace_size
include/flux/cuda/gemm_impls/gemm_operator_base_default_impl.hpp:78
Method
get_world
(self)
python/flux/dist_utils.py:61
Function
get_world_size_from_env
src/cuda/utils.cc:83
Method
global_rank_to_local_rank
include/flux/utils.h:54
Method
global_red
include/flux/cuda/memory_utils.hpp:64
Method
global_red
include/flux/cuda/memory_utils.hpp:80
Method
global_red
include/flux/cuda/memory_utils.hpp:99
Method
global_red
include/flux/cuda/memory_utils.hpp:116
Function
has_heterogeneous_nvlink
src/ths_op/topo_utils.cc:326
Function
has_heterogeneous_pcie
src/ths_op/topo_utils.cc:347
Method
has_nvlink
src/reduce_scatter/ths_op/gemm_reduce_scatter.cc:160
Function
has_nvswitch
src/ths_op/topo_utils.cc:319
Method
if
src/reduce_scatter/epilogue_vectorized_reduce_scatter.hpp:281
Method
if
src/reduce_scatter/epilogue_nvshmem_reduce_scatter.hpp:306
Function
index_seq_split_slice
include/flux/flux.h:371
Function
init_ag_kernel_ops
src/all_gather/ths_op/all_gather_gemm_kernel.cc:919
Method
init_dp_tile_work
src/all_gather/sm80_all_gather_gemm.hpp:627
Method
init_dp_tile_work
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:600
Function
init_flux_shm
src/ths_op/flux_shm.cc:158
Method
init_iterator_A
src/all_gather/sm80_all_gather_gemm.hpp:579
Method
init_iterator_A
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:528
Method
init_iterator_B
src/all_gather/sm80_all_gather_gemm.hpp:604
Method
init_iterator_B
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:577
Function
init_local_groups
()
test/utils.py:65
Method
init_sk_tile_work
src/all_gather/sm80_all_gather_gemm.hpp:651
Method
init_sk_tile_work
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:624
Function
init_with_c10d_pg
pynvshmem/src/functions.cpp:26
Method
init_workspace
Assign and initialize the specified workspace buffer. Assumes the memory allocated to workspace is at least as large as get_workspace_size().
src/all_gather/sm80_all_gather_gemm.hpp:431
Method
init_workspace
Assign and initialize the specified workspace buffer. Assumes the memory allocated to workspace is at least as large as get_workspace_size().
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:382
Method
initialize
Initializes GEMM state from arguments and workspace memory
src/reduce_scatter/gemm_v2_reduce_scatter.hpp:60
Method
initialize_args_workspace
src/reduce_scatter/gemm_v2_reduce_scatter.hpp:477
Function
initialize_inter_node_rs_group
(tp_group: dist.ProcessGroup, nnodes: int)
python/flux/gemm_rs_sm80.py:60
Method
initialize_workspace
src/all_gather/sm90_all_gather_gemm_tma_warpspecialized_cooperative.hpp:323
Method
initialize_workspace
src/reduce_scatter/sm90_epilogue_evt.hpp:154
Method
initialize_workspace
src/reduce_scatter/sm90_gemm_tma_warpspecialized_cooperative_reduce_scatter.hpp:271
Method
invoke
Factory invocation
src/all_gather/sm80_all_gather_gemm.hpp:1008
Method
invoke
src/reduce_scatter/reduce_scatter_kernel.hpp:626
Method
invoke
src/reduce_scatter/reduce_scatter_kernel.hpp:802
Method
invoke
src/reduce_scatter/reduce_scatter_kernel.hpp:923
Method
invoke
src/reduce_scatter/reduce_scatter_kernel.hpp:1066
Method
invoke
src/reduce_scatter/reduce_scatter_kernel.hpp:1217
Method
invoke
src/reduce_scatter/reduce_scatter_kernel.hpp:1305
Method
invoke
src/reduce_scatter/reduce_scatter_kernel.hpp:1435
Method
invoke
Factory invocation
src/reduce_scatter/gemmk_universal_with_visitor_streamk.h:943
← previous
next →
501–600 of 784, ranked by callers